Disaster Resilience of Urban Communities in Canada: New Probabilistic Models and Computational Methods

点击翻译按钮获取中文摘要

An earthquake in the Metro Vancouver region can cause extensive damage and displacement of a significant portion of the population, but we lack knowledge about how the built environment will perform. A recent report by the British Columbia Auditor General concluded that the region is not adequately prepared for a catastrophic earthquake. Studies have shown that a strong earthquake near Vancouver, or an offshore subduction earthquake, is likely to cause billions of dollars in damage, but the impacts are uncertain. The proposed work employs that region as a testbed for a new computer simulation platform with interchangeable probabilistic models aimed at helping the region better prepare for and recover from a disaster. Rather than seeing a computer simulation program as a single deliverable from this work, an important outcome is a collection of probabilistic models for damage and recovery of bridges, ports, residential houses, and structures in the power and water grids. These models can be shared with other researchers, and models from other researchers are integrated into the framework. The information flowing between the models include calculated responses, such as earthquake intensity, and flow of power, water, and resources for repairs. The computational platform has already been employed to simulate regional losses due to earthquakes and the flow of fuel to coastal communities. A central question addressed in the proposed work is how many persons will be displaced in the next strong earthquake near Vancouver and what the timeline is for the housing recovery. Additional performance and resilience metrics are predicted and "what-if" scenarios are simulated, but population displacement and long-term housing recovery represent important gaps in our understanding of community resilience. Many models are needed to address that problem and a framework of hazard models are already in place from earlier work. The proposed work creates a new framework of probabilistic models in three categories: 1) Models for damage to residences and the structures of the power/water grids, directly affecting the number of displaced persons; 2) Models to predict the repair actions and the resources needed for repairs, without which housing recovery will be delayed; and 3) Models for damage and recovery of port and bridge structures critical to the flow of workers and materials into and within the region. The United Nations framework for disaster risk reduction urges governments to reduce the damage potential for critical infrastructure, and the Ministry of Public Safety in Canada has a mandate to develop plans to enhance the disaster resilience of Canadian communities. The proposed work addresses those challenges through the development of new tools and the training of highly qualified personnel to help managers of ports, bridges, and power and water infrastructure to assess and to reduce the risks associated with the structures they oversee.